In a case brought before the Lyon CAA by our firm, the tax judge reiterated his commitment to the median of an arm's length interval for assessing an indirect profit transfer. In this case, the company's margin fell below the first quartile of an interval made up of some thirty references. This was all it took for the tax authorities to consider that a transfer pricing anomaly was affecting the company's profitability, and consequently to adjust its net margin to the median of the panel.
The Lyon Administrative Court endorsed this approach, repeating almost verbatim the operative part of the GE Medical system ruling and considering that "in the circumstances of the case, the median [...], which makes it possible to limit the margins of approximation in relation to a point situated at one or other of the extreme limits of this interval, must be regarded as the point of the interval which best reflects the facts and circumstances of the transactions concerned" (Lyon Administrative Court, June 22, 2021, N°1909917 and 1910206).
This recital continues to leave us stunned, as these two decisions are conspicuous by their failure to explain the "facts and circumstances" apparently so exceptional as to justify the administration's automatic assumption that the taxpayer must necessarily generate a higher margin than half of the companies deemed comparable. We therefore expressly asked the Lyon CAA to take a firm and explicit stance on this issue, both in our writings and at the hearing, in order to provide the clarity needed to ensure legal and tax certainty for taxpayers. However, the Lyon CAA noted that "the investigation revealed that the net margins generated on business flows with companies X and Y were not only lower than the lowest of the data in the arm's length interval, but also lower than the "extreme" minima of the sample observed. For its part, the claimant merely points out that the margins making up the first quartiles already place it above six companies in the two panels for 2010 and 2011, and that it is not clear from the investigation that the administration applied the median of the interquartile range automatically and without any assessment of the circumstances of the transactions concerned. The claimant, for its part, does not justify any specific circumstances making it possible to establish that the administration, in view of the transactions in dispute, should have deviated from this median margin".
- We deduce the following from this decision:
The French tax authorities are confirmed in their automatic reference to the median of arm's length panels, whenever the taxpayer's remuneration deviates from this median. - It is up to the taxpayer to provide any factual, statistical, economic or other evidence to show that the tax authorities should have deviated from this median, given the transactions in dispute. In practice, this reverses the burden of proof onto the taxpayer, forcing him to produce a highly subjective analysis.
The Court's recital does, however, seem to contain a positive clue: if the taxpayer's margin already falls within the arm's length (interquartile) range, the system could be reversed, as the famous "facts and circumstances" would be different. Indeed, it was because the margin fell outside the range that the Court had to retain a target.